Research Flashcards
3 factors involved in a evidence based practice
Expertise
Evidence
Patients
Type of research that stablish cause and effect relationship
True experimental research
Criterias of true experimental group
Controlled and the experimental group
Researcher manipulated group
Randominization
If one or more of the criteria of true experimental research is not met.
Quasi experimental research
Most commonly not attained criteria to fall under Quasi experimental research
Randominization
Type of research which describes characteristics, attitude, behavior, beliefs, manifestations
Non experimental descriptive
Type of research which describes connections or relationships between variables and it doesn’t manipulate parients
Non experimental exploratory
Type of research design that uses collection of several researches
Synthesized literature
2 types of synthesized literature
Systematic reviews
Meta analysis
Type of synthesized literature that focuses on comprehensive analysis of FULL RANGE LITERATURE
Systhematic reviews
Type of synthesized literature that uses statistical combination of several researches
Higher level if validity
Meta analysis
2 types of Qualitative research designs
Phenomenology
Ethnography
Type of Qualitative research design that focuses on specific events
Phenomenology
Type of Qualitative research design that focuses on ethnicity
Describes culture, traditions, attitude, and beliefs
+SITE VISIT
It relates health conditions with their ethnics
Ethnography
6 types of Quantitative research designs
Randomized controlled trial (RCT) Cohort/follow up study Case control Cross sectional Longitudinal Singke subject design
Gold standard of all experimental researches
High validity
A true experimental research
Randomized controlled trial
Prospective study
Subjects may NOT HAVE the disorder yet
Waits for the future results
Cohort/Follow-up study
Retrospective study
Subjects may already HAVE the disorder
Case control
Studies group of subjects at specific point in time
Cross sectional
Studies SERIES of cohort studies or follow up studies
Example: milestones
Longitudinal
Studies effectiveness of a specific intervention
Single subject design
Single subject design for effectiveness of a specific intervention
A-B
Single subject design for the duration of effectiveness
+wash out period(no tx)
A-B-A
Single subject design
if the independent variable is the direct cause of change in the dependent variable.
Identifying if there is a intervening variables
A-B-A-B
Building blocks of a research title or research question
Variables
Cause
Makes change
Intervention or treatment
AKA predictor variable
Independent variable
Effect
The one that will change
AKA target variable
Dependent variable
Decreases validity of the research
AKA unaccounted variable/covariable
Intervening variable
Educated guess
Hypothesis
A hypothesis that does not expect/assume
More on negative
Null hypothesis
A hypothesis that expects/assume
More on positive
Alternative hypothesis
Type of error when the researcher rejects the null when its in fact it is true
Implications: giving a tx that is NOT REALLY EFFECTIVE
Type I error
Type of error when the researcher accepts the null when it is false
Implications: ignoring a tx that is REALLY EFFECTIVE
Type II error
Weakest or lowest data type
Ex. Gender, blood type, course
AKA classificatory scale
Nominal data
Rank ordered category
Every rank has its own description
Ex. Mmt grade, pain scale, fim, borg scale(Likert scale)
Ordinal data
Rank order with known distance
No absolute Zero(can extend beyond zero)
Ex. Temperature
Interval data
True absolute Zero
Highest level
Zero means nothing
Ex. Height, weight,age, length
Ratio
Representative of the population
Sample
2 types of sampling
Probability sampling technique
Non probability sampling technique
4 types of probability sampling
Simple random sampling
Stratified random sampling
Systematic sampling
Cluster sampling
Classic fishbowl technique
All have EQUAL chance of being a part of the sample
Simple random sampling
+Layers/group
All have chance but UNEQUAL
Divided according to characteristics
Stratified random sampling
Sampling interval
Every _nth
UNEQUAL chance
Systematic sampling
For large populations
UNEQUAL chance
Divided into groups regardless of characteristics
Cluster sampling
4 types of Non probability sampling
Convinience sampling
Purposive sampling
Snowball sampling
Quota sampling
Sample is chosen based on availability
Convinience sampling
+criteria
Purposive sampling
Criteria-> chain referral
Snow ball sampling
Proportions
Quota sampling
Refers to Correctness, appropriateness
Validity
2 types of validity
Internal
External
Independent variable is the direct cause of change in dependent variables with NO intervening variable
Ex. US
Internal validity
Ability to generalize the research to the general population with same characteristics
Ex. Intervention is applied to other patients =same effects
External validity
Threats to internal validity
Hawthorne effect
Placebo effect
Carry over effect
A threat to internal validity the Tendency to perform better because they are being observed
Hawthorne effect
A threat to internal validity the Tendency to report better due to a sham tx(false)
Purely psychological
Placebo effect
A threat to internal validity Due to multiple tx intervention
Many intervening variables
Carry over effect
Threats to external validity
Interaction of tx and selection
Interaction of tx and setting
Interaction of tx and history
Refers to consistency, reproducibility, free from error
Reliability
Reliability of 2 or more raters in a single trial
Intertester reliability
Reliability of 1 rater in a multiple trials
Intratester reliability
Reliability from the stability of the outcome measure
Test-retest reliability
Ability to obtain a positive result when the condition is present
True positive
Ability to obtain a negative result when the condition is really absent
True negative
If the test has high ______ a NEGATIVE result RULE OUT diagnosis
Sensitivity
If a test has high ______ a POSITIVE result RULES IN the diagnosis
Sensitivity
Measures of central tendency
Mean
Median
Mode
Average score
Mean
Divides the scores into 2 halves
Median
Most frequent score in a distribution
Mode
Measures of variables
Range
Standard
Variance
Difference of the highest and lowest scores
Range
Difference around the mean
Standard deviation
Difference among the scores
Varience
Normal distribution bell curve
Mesokurtic curve
Far apart bell curve
Platykurtic curve
Bell curve that has a concentrated distribution on one side
Leptokurtic curve
Tail to the right
Majority are below average
Positively skewed
Tail to the left
Majority are below average
Negatively skewed
Ratio and interval
Parametric
Nominal and ordinal
Nonparametric
2 types of transformation tests
Brush
Rasch
Converts interval to ordinal
Brush
Code: BIO ROI
Converts ordinal to interval
Rasch
Code: BIO ROI
Called the predictor variable
Independent variable
Ability to generalize the research to the general population with the same characteristics
External validity
To meet the definition of reliable a test must be both what
Consistent
Free from error
To compare a distribution
Pie graph
To illustrate a change in the variables
Line graph
To illustrate a larger change in the variables
Bar graph
If you want to graphically illustrate the relationship of two variables.Which type of graph would you select?
Scatter plot
A px uses a self administered assessment tool as a method to record daily progress. What type of reliability?
Intratester
